RESUMO
OBJECTIVE: Current surveillance for healthcare-associated (HA) urinary tract infection (UTI) is focused on catheter-associated infection with hospital onset (HO-CAUTI), yet this surveillance does not represent the full burden of HA-UTI to patients. Our objective was to measure the incidence of potentially HA, community-onset (CO) UTI in a retrospective cohort of hospitalized patients. DESIGN: Retrospective cohort study. SETTING: Academic, quaternary care, referral center. PATIENTS: Hospitalized adults at risk for HA-UTI from May 2009 to December 2011 were included. METHODS: Patients who did not experience a UTI during the index hospitalization were followed for 30 days post discharge to identify cases of potentially HA-CO UTI. RESULTS: We identified 3,273 patients at risk for potentially HA-CO UTI. The incidence of HA-CO UTI in the 30 days post discharge was 29.8 per 1,000 patients. Independent risk factors of HA-CO UTI included paraplegia or quadriplegia (adjusted odds ratio [aOR], 4.6; 95% confidence interval [CI], 1.2-18.0), indwelling catheter during index hospitalization (aOR, 1.5; 95% CI, 1.0-2.3), prior piperacillin-tazobactam prescription (aOR, 2.3; 95% CI, 1.1-4.5), prior penicillin class prescription (aOR, 1.7; 95% CI, 1.0-2.8), and private insurance (aOR, 0.6; 95% CI, 0.4-0.9). CONCLUSIONS: HA-CO UTI may be common within 30 days following hospital discharge. These data suggest that surveillance efforts may need to be expanded to capture the full burden to patients and better inform antibiotic prescribing decisions for patients with a history of hospitalization.